DESC:FORM 2

THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003

COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of invention:
DATA ANALYTICS SYSTEM AND A METHOD OF PROCESSING THEREOF

Applicant:
Legasis Private Limited
A company Incorporated in India having address:
12A/09, 13TH FLOOR, PARINEE CRESENZO,
G BLOCK, BKC, BANDRA EAST,
Mumbai - 400051, MAHARASHTRA, India

The following specification particularly describes the invention and the manner in which it is to be performed.

CROSS REFERENCE TO RELATED APPLICATION AND PRIORITY
[0001] The present invention claims priority from Indian patent application 202321067528 filed on date 9th October 2023.

FIELD OF THE INVENTION
[0002] The present disclosure generally relates to an AI-based legal data analytics and optimization system, along with a method for processing legal data. Specifically, this disclosure pertains to an artificial intelligence-driven system designed to extract and provide meaningful legal insights from a dataset that may not be readily apparent to human researchers. This innovative approach leverages advanced AI techniques to analyze complex legal information, enabling users to gain a deeper understanding of legal trends, patterns, and implications that can inform their decision-making processes.

BACKGROUND OF THE INVENTION
[0003] In legal research, precedents, plaints, petitions, and other legal documents are essential for lawyers and concerned individuals. These documents contain crucial elements, such as judicial interpretations, legal statutes with accompanying facts, and analytical insights into relevant legislation, which can significantly influence legal strategy in subsequent cases.
[0004] Traditionally, legal research platforms associate metadata—such as keywords and phrases—with a legal database to retrieve relevant judgments or documents. However, this process can yield hundreds of results related to a user's case, making it challenging to identify which judgments are pertinent and in what order they should be prioritized. This sorting is often a tedious and complex task.
[0005] Conventional legal research methodologies and platforms have several limitations, including time consumption, slow processing speeds, steep learning curves for inputting variables, and the need for multiple inputs. Specifically, extracting key data points from previously decided cases using these traditional methods is time-consuming and prone to errors. The manual nature of this approach relies heavily on an individual’s ability to discern various parameters within a judgment and apply them to the selection process, resulting in an unreliable, slow, and inaccurate legal research experience.

OBJECTS OF THE INVENTION
[0006] An object of the present disclosure is to provide an AI based system facilitating data extraction and a method thereof.
[0007] Another object of the present disclosure is to provide predictive analytics, automation of repetitive tasks, and a personalized experience to the individuals.
[0008] Yet another object of the present disclosure is to provide extraction of meaningful legal insights from the judgements that may not be immediately apparent to human researchers/ the individuals. This could include identifying patterns in legal decisions, highlighting key legal principles, or summarizing complex legal arguments.
[0009] Yet another object of the present disclosure is to provide analysis of how specific judgements have been cited and used as legal precedents in subsequent cases. This can help the individuals understand the impact and significance of particular judgements over time.
[0010] Yet another object of the present disclosure is to provide prediction of how a specific legal judgement may influence future legal decisions. This can be valuable for the individuals in assessing the potential implications of a judgement.
[0011] Yet another object of the present disclosure is to provide enhancement of search capabilities of legal research tool by implementing advanced natural language processing (NLP) powered by Open Artificial Intelligence (AI's) Application Programming Interface (APIs). This enables the individuals to ask complex legal questions in plain language, and the tool could provide relevant case law and explanations.
[0012] Yet another object of the present disclosure is to provide automatic generation of concise and informative summaries of legal judgements using NLP. These summaries can save legal researcher/ individual’s significant time and effort in reviewing lengthy judgements.
[0013] Yet another object of the present disclosure is to provide a dashboard that tracks and visualizes trends in legal decisions over time. This could include graphical representations of how specific legal issues have evolved in the courts.
[0014] Yet another object of the present disclosure is to provide legal argument generator for assisting legal professionals in generating persuasive legal arguments based on the content of legal judgements. This can be a valuable tool for the individuals in preparing cases.
[0015] Yet another object of the present disclosure is to provide a risk assessment module that evaluates the potential legal risks associated with specific judgements or legal strategies. This can assist in making informed decisions.
[0016] Yet another object of the present disclosure is to provide analysis of the individual’s research history and preferences to provide personalized recommendations for relevant legal judgements and research materials.

SUMMARY OF THE INVENTION
[0017] Before the present open-source AI-based system facilitating data extraction and a method thereof are described, it is to be understood that this application is not limited to the particular system and methodologies described, as there can be multiple possible embodiments that are not expressly illustrated in the present disclosure. It is also to be understood that the terminology used in the description is to describe the particular implementations, versions, or embodiments only and is not intended to limit the scope of the present application.
[0018] In an embodiment, the present disclosure provides a system wherein the processing circuitry (106) further sorts judgments that have been referred to in subsequent rulings based on: the hierarchy of the courts, giving preference to higher courts; and the timeline of the judgments, prioritizing more recent rulings.
[0019] In yet another embodiment, the system provides that the processing circuitry (106) is further configured to cross-verify judgments to determine if they have been overturned in subsequent rulings.
[0020] In still another embodiment, the present disclosure provides a system wherein the system provides automated summaries of complex legal judgments using NLP algorithms to distill long rulings into concise, informative summaries.
[0021] In an embodiment, the present disclosure provides a system wherein the system provides predictive analytics by analyzing legal trends and patterns to evaluate how a particular judgment may influence future decisions.
[0022] In yet another embodiment, the system provides that the system supports natural language search capabilities, allowing users to input complex legal questions in plain language and receive relevant case laws and insights.
[0023] In yet another embodiment, the processing circuitry is configured with a machine-learning model configured to perform predictive analytics, thereby predicting specific legal judgments that may influence future legal decisions.
[0024] In yet another embodiment, the system provides that the processing circuitry is configured to analyze the selected judgments to identify patterns in legal decisions, key legal principles, and to summarize complex legal arguments, including relevant facts, legal issues, reasoning provided, reasons cited by the court, and references to specific acts and section numbers.
[0025] The present method for legal data extraction comprises several key steps. It begins with receiving a query from a user via a user interface, wherein the query includes at least one legal issue. Next, natural language processing (NLP) models are utilized to extract a plurality of facts and legal issues from the received query, allowing for analysis in both factual and legal contexts. Following this, relevant legal judgments are identified from a database, ensuring that these judgments relate to at least one of the extracted legal issues. Each identified judgment is assigned a Fact Relevancy Score (FRS) based on the factual proximity between the query and the judgment, which serves to filter out any irrelevant judgments. The identified legal judgments are then sorted based on their proximity to the plurality of legal issues using a BM25 ranking algorithm, with each judgment receiving a score based on the proximity evaluated through the frequency of specific words, phrases, or legal contexts present in the query. The method includes verifying whether the identified judgments have been referred to or dissented in subsequent rulings, followed by further sorting of the identified judgments based on court hierarchy and the timeline of the judgments, prioritizing higher court rulings and more recent judgments. Finally, the method retrieves relevant case laws based on their proximity to the extracted legal issues and their timeline. Additionally, the method encompasses presenting the retrieved relevant case laws to the user in a user-friendly format, providing automated summaries of complex legal judgments through NLP algorithms to distill lengthy rulings into concise and informative summaries, and enabling the user to perform natural language searches by inputting complex legal questions in plain language to receive pertinent case laws and insights.
[0026] In an embodiment, the present invention presents the retrieved relevant case laws to the user in a user-friendly format. Additionally, it includes providing automated summaries of complex legal judgments by utilizing natural language processing (NLP) algorithms to distill lengthy rulings into concise and informative summaries. The method also enables the user to perform natural language searches by inputting complex legal questions in plain language and receiving relevant case laws and insights.
[0027] In yet another embodiment, the present invention provides the steps of sorting and retrieving relevant legal judgments further comprise applying a cross-verification step to determine whether any identified judgments have been overturned in subsequent rulings. This embodiment also includes analyzing legal trends and patterns to evaluate how particular judgments may influence future decisions through predictive analytics.
BRIEF DESCRIPTION OF DRAWINGS
[0028] The foregoing detailed description of embodiments is better understood when read in conjunction with the appended drawings. For illustrating the present subject matter, an example of construction of the present subject matter is provided as figures; however, the present subject matter is not limited to the specific system and the method disclosed in the document and the figures.
[0029] Figure 1 illustrates a block diagram of a system (100) for facilitating a legal data extraction, in accordance with an embodiment of the present subject matter.
[0030] Figure 2 illustrates a flow diagram of a method (200) for facilitating data extraction, in accordance with an embodiment of the present subject matter.

DETAILED DESCRIPTION OF THE INVENTION
[0031] Some embodiments of the present disclosure, illustrating all its features, will now be discussed in detail. The words “comprising/comprises”, and other forms thereof, are intended to be equivalent in meaning and be open-ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms "a", "an" and "the" include plural references unless the context clearly dictates otherwise. Although any systems and methods similar or equivalent to those described herein can be used in the practice or testing of embodiments of the present disclosure, the exemplary, systems and methods facilitating data extraction are now described. The disclosed embodiments of the systems and the methods facilitating data extraction are merely exemplary of the disclosure, which may be embodied in various forms.
[0032] The system (100) comprises a user interface (102), a storage device (104), and processing circuitry (106). The storage device (104) stores a plurality of instructions to be executed by the processing circuitry (106). The system (100) may be configured as a platform for internal and external deployment in an organization or industry. For example, the system (100) herein will be explained for a legal organization. The system (100) leverages advanced technologies like artificial intelligence (AI) and machine learning (ML) for legal research, automating processes such as judgment ranking, predictive analytics, and personalized legal insights.
[0033] Referring now to Figure 1, the system (100) is illustrated in accordance with an embodiment of the present subject matter. The system (100) comprises a user interface (102), a storage device (104), and at least one processing circuitry (106). The processing circuitry (106) may include microprocessors, microcontrollers, or any other devices that manipulate signals based on operational instructions. Among other capabilities, the processing circuitry (106) is configured to fetch and execute computer-readable instructions stored in the storage device (104), enabling features like query ingestion, legal issue extraction, and judgment sorting.
[0034] The storage device (104) may include any computer-readable medium known in the art, such as volatile memory like SRAM and DRAM, and non-volatile memory like ROM, flash memories, hard disks, and optical disks. It stores essential instructions for executing complex legal data extraction tasks.
[0035] Referring now to Figure 1, a block diagram of a system (100) for facilitating legal data extraction is illustrated. The system (100) enables the generation of relevant case laws. A user interface (102) allows users to upload legal queries, such as complaints or legal issues. The system (100) processes these inputs using AI-powered natural language processing (NLP) models. The system extracts facts and legal issues from uploaded queries and analyzes them with the help of AI models and customizable prompts. These prompts, which are user-engineered, guide the AI throughout various steps, such as identifying judgments, extracting facts, and sorting relevant case laws.
[0036] In one embodiment, the received query may contain multiple legal issues, each of which is marked for relevance. The system then analyzes case laws, opinions, or rulings of judges corresponding to each legal issue. The system’s BM25 ranking algorithm evaluates the relevance of judgments in its database by assessing their proximity to the extracted facts and legal issues.
[0037] The system (100) uses a comprehensive judgment database for retrieving and ranking judgments. The NLP module enables the system to analyze queries based on their context and text, breaking down complex legal language into digestible segments. These segments are cross-referenced with existing legal judgments, which are further sorted and scored for factual proximity and legal relevance.
[0038] The BM25 ranking algorithm helps determine how closely the retrieved judgments match the user’s query. The proximity of judgments is gauged by the frequency of specific words, phrases, or legal context within the query. The system sorts judgments from the highest to lowest relevancy, filtering out dissented judgments that might undermine legal validity.
[0039] The system then scores and ranks each judgment based on its factual relevance, using a Fact Relevancy Score (FRS). Judgments are further sorted based on the hierarchy of courts, with preference given to Supreme Court rulings, followed by High Court, Appellate Tribunals, and other lower courts. The system retrieves judgments from each level, ranking them based on both proximity and court hierarchy.
Additionally, judgments are organized by timeliness, ensuring that more recent judgments are prioritized. The system identifies the latest rulings related to the query, offering legal professionals the most up-to-date case laws. For instance, the system might retrieve a combination of Supreme Court and High Court rulings for a specific issue, ranking them based on their legal importance and decency.
[0040] Further, the system (100) retrieves the top 5 relevant case law/judgments based on their proximity and timeline. The system processes user queries by analyzing the query in both factual and legal contexts, extracting key legal issues and correlating them with previous judgments.
[0041] The system (100) provides automated insights into legal precedents by evaluating how certain judgments have been used in subsequent cases, analyzing their impact on future legal decisions through ML models. This predictive feature allows users to assess potential outcomes based on historical data and patterns.
[0042] Additionally, the system (100) provides legal professionals with automated summaries of complex legal judgments, saving time by distilling long rulings into concise, informative summaries using NLP algorithms. The system’s AI capabilities, powered by Open AI’s APIs, further enhance its ability to generate these summaries.
[0043] The system (100) also incorporates predictive analytics, evaluating how a particular judgment could influence future decisions. By analyzing legal trends and patterns, the system offers a powerful tool for attorneys and researchers to anticipate potential legal outcomes.
[0044] In another embodiment, the system (100) supports natural language search capabilities, allowing users to ask complex legal questions in plain language and receive relevant case law and legal insights, enhancing research efficiency.
[0045] To improve research productivity further, the system (100) provides personalized legal research recommendations based on user preferences and history, suggesting relevant judgments and insights based on the user's previous research.
[0046] Figure 2 illustrates a flowchart depicting a method (200) for legal data extraction, in accordance with an embodiment of the present subject matter. The sequence in which the method 200 is described is not meant to be limiting, and any number of the described method steps may be combined in any order to implement method 200 or alternative methods. Additionally, individual steps may be omitted from method 200 without departing from the spirit and scope of the subject matter described herein. Furthermore, the method may be implemented in any suitable form, including hardware, software, firmware, or a combination thereof. However, for ease of explanation, the following embodiments describe method 200 as implemented within system 100.
[0047] At block 202, receiving a query from a user via a user interface, wherein the query includes at least one legal issue.
[0048] At block 204, utilizing natural language processing (NLP) models to extract a plurality of facts and legal issues from the received query and to analyze the query in both factual and legal contexts
[0049] At block 206, identifying relevant legal judgments from a database, wherein the identified legal judgments relate to at least one of the extracted legal issues.
[0050] At block 208, assigning a Fact Relevancy Score (FRS) to each identified judgment based on the factual proximity between the query and the judgment, thereby filtering out irrelevant judgments.
[0051] At block 210, sorting the identified legal judgments based on their proximity to the plurality of legal issues using a BM25 ranking algorithm, wherein each judgment is assigned a score based on the proximity evaluated through the frequency of specific words, phrases, or legal context within the query.
[0052] At block 212, verifying whether the identified judgments have been referred to or dissented in subsequent judgments. At block 214, sorting the identified judgments based on court hierarchy and the timeline of the judgments, prioritizing higher court rulings and more recent judgments. At block 216, retrieving the relevant case laws based on their proximity to the extracted legal issues and their timeline.
[0053] The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the spirit and scope of the embodiments as described herein.
,CLAIMS:
1. A system for legal data extraction, comprising:
a user interface (102) configured to receive a query from a user, wherein the query includes at least one legal issue;
a storage device (104) configured to store a plurality of instructions and a database of legal judgments;
a processing circuitry (106) coupled with the storage device configured to:
use natural language processing (NLP) models to extract a plurality of facts and legal issues from the received query and to analyze the query in both factual and legal contexts;
identify relevant legal judgments from the database, wherein the identified legal judgments relate to at least one of the extracted legal issues;
assign a Fact Relevancy Score (FRS) to each identified judgment based on the factual proximity between the query and the judgment, filtering out irrelevant judgments;
sort the identified legal judgments based on their proximity to the plurality of legal issues using a BM25 ranking algorithm, wherein each judgment is assigned a score based on the proximity evaluated through the frequency of specific words, phrases, or legal context within the query;
verify whether the identified judgments have been referred to or dissented in subsequent judgments;
further sort the identified judgments based on court hierarchy and the timeline of the judgments, prioritizing higher court rulings and more recent judgments;
retrieve the relevant case laws based on their proximity to the extracted legal issues and their timeline.
2. The system of claim 1, wherein the processing circuitry (106) further sorts judgments that have been referred to in subsequent rulings based on:
• the hierarchy of the courts, giving preference to higher courts; and
• the timeline of the judgments, prioritizing more recent rulings.
3. The system of claim 1, wherein the processing circuitry (106) is further configured to cross-verify judgments to determine if they have been overturned in subsequent rulings.

4. The system of claim 1, wherein the system provides automated summaries of complex legal judgments using NLP algorithms to distill long rulings into concise, informative summaries.

5. The system of claim 1, wherein the processing circuitry is configured with a machine learning model configured to perform predictive analytics, thereby predicting specific legal judgments that may influence future legal decisions.

6. The system of claim 1, wherein the system supports natural language search capabilities, allowing users to input complex legal questions in plain language and receive relevant case laws and insights.

7. The system of claim 1, wherein the processing circuitry is configured to analyze the selected judgments to identify patterns in legal decisions, key legal principles, and to summarize complex legal arguments, including relevant facts, legal issues, reasoning provided, reasons cited by the court, and references to specific acts and section numbers.

8. A method for legal data extraction, comprising:
receiving a query from a user via a user interface, wherein the query includes at least one legal issue;
utilizing natural language processing (NLP) models to extract a plurality of facts and legal issues from the received query and to analyze the query in both factual and legal contexts;
identifying relevant legal judgments from a database, wherein the identified legal judgments relate to at least one of the extracted legal issues;
assigning a Fact Relevancy Score (FRS) to each identified judgment based on the factual proximity between the query and the judgment, thereby filtering out irrelevant judgments;
sorting the identified legal judgments based on their proximity to the plurality of legal issues using a BM25 ranking algorithm, wherein each judgment is assigned a score based on the proximity evaluated through the frequency of specific words, phrases, or legal context within the query;
verifying whether the identified judgments have been referred to or dissented in subsequent judgments;
further sorting the identified judgments based on court hierarchy and the timeline of the judgments, prioritizing higher court rulings and more recent judgments;
retrieving the relevant case laws based on their proximity to the extracted legal issues and their timeline.
9. The method of claim 1, further comprising:
presenting the retrieved relevant case laws to the user in a user-friendly format;
providing automated summaries of complex legal judgments using NLP algorithms to distil long rulings into concise, informative summaries;
enabling the user to perform natural language searches by inputting complex legal questions in plain language and receiving relevant case laws and insights.
10. The method of claim 1, wherein the steps of sorting and retrieving relevant legal judgments further comprise:
applying a cross-verification step to determine if any identified judgments have been overturned in subsequent rulings;
analysing legal trends and patterns to evaluate how particular judgments may influence future decisions through predictive analytics.